Measurement uncertainty and conformity assessment - From uncertainty to risk

The “Guide to the Expression of Uncertainty in Measurement” (GUM) provides a general framework for evaluating and expressing measurement uncertainty, applicable to various fields such as health, energy, commerce, and the environment. These rules harmonize practices and ensure the reliability of measurements, whether they involve medical tests to diagnose a disease or adjust treatment, energy billing, the detection of micropollutants in water, or real-time monitoring of air quality in urban areas.

Measurement uncertainty characterizes the quality of results, enabling stakeholders to interpret them and make informed decisions.

Measurement uncertainty is the doubt regarding the “true value” of the measurand that remains after measurements have been taken. Measurement uncertainty can be expressed in the following ways:

• a standard uncertainty;

• an expanded uncertainty;

• a coverage interval;

• a probability density.

The measurement result, which includes the measured value and the associated uncertainty, must be presented in a form that is understandable and usable for the specific purpose, and must include all available information.

Without uncertainty, a measurement result can no longer be compared:

• with another measurement result;

• to a reference value specified in a standard;

• to a specification.

Based on this information, the end user makes decisions: acceptance or rejection of a hypothesis in research and development, control of a manufacturing process, acceptance or rejection of a product, medical diagnosis, environmental protection, etc.

Interlaboratory comparisons can also play a crucial role in estimating measurement uncertainty, especially in situations where it is difficult to establish a measurement model or when the instrumentation used has limitations. They can, for example, help identify biases or missing sources of uncertainty that would not be easy to detect in an isolated laboratory.

This article is organized into four main chapters:

• Section 1